Discovering Properties¶

Source distressed and foreclosure properties in a growth area, then move them into your pipeline for screening.

Prerequisites¶

A growth area to search in. Run the Growth Explorer for a state first, or pick an existing area from the list. Without one, discovery shows a market picker instead.
API keys for the sources you want to use (optional but recommended) — see Data Sources & API Keys.

Step 1: Choose a Growth Area¶

From the Growth Areas list, click Discover Properties on a market row — this opens /discovery/?growth_area_id=X pre-filtered to that market.
Or open /discovery/ directly: you'll see a picker of your top growth areas.

Step 2: Review Available Sources¶

The discovery page lists the available sources for the chosen market, with a count of existing records per source:

Source	What it contains	Has rent data?
HUD REO	Government-owned HUD foreclosures	No
USDA REO	USDA Rural Development foreclosures	No
VRM (VA REO)	VA-owned foreclosures via VRM	Yes (projected rent)
ATTOM Pre-foreclosure	NOD/NTS pre-foreclosure notices	No
County Foreclosure Notices	County-level NTS / sheriff sale / auction records	No

Rent data matters for screening. Properties from VRM carry projected rents, so gross-yield and price-to-rent screening can run. HUD, USDA, ATTOM, and County sources have no rent estimate — yield and PTR are skipped for them (see Screening Properties).

Each source shows its count — how many matching records already exist in the app for that state + city. Check the boxes for the sources you want to query.

Step 3: Run Discovery¶

Select at least one source (you'll get a warning if none are selected).
Click Discover Properties in [City].

What happens:

prei queries the selected sources for the growth area's state + city and creates PipelineProperty records for matches.
VRM scrape runs in a background thread if no VA listings exist for the state yet (Gunicorn worker timeout is 30s) — you'll see an info message telling you to refresh in a minute. HUD/USDA bulk data loads run synchronously with retry-on-lock handling.
When finished, results render in-page on the discovery page — you stay in context instead of being redirected.

Step 4: Review Results¶

Discovery shows the results right on the page:

KPI cards — discovered, passed screening, failed, and already-existing properties in this market.
Per-source results — what was found for each selected source.
Success/warning messages at the top describing what was created.
A View in Screener → button to jump to /pipeline/screener/?growth_area_id=X and review the properties in detail.

Continue to Screening Properties.

Source Details¶

HUD REO¶

Government-owned foreclosures sourced from the HUD API. Good volume of banked properties; no rent estimates, so yield/PTR screening is skipped.

USDA REO¶

Rural Development foreclosures, similar shape to HUD. Relevant when you invest in smaller towns that qualify for USDA programs.

VRM (VA REO)¶

VA-owned foreclosures via VRM Properties. The only source with projected monthly rent, enabling full yield/PTR screening.

ATTOM Pre-foreclosure¶

Notice of Default (NOD) and Notice of Trustee Sale (NTS) records. Properties here are still in pre-foreclosure — earlier stage, more upside, more uncertainty.

County Foreclosure Notices¶

NTS, sheriff sale, and auction records at the county level. Often the earliest notice of distress in a market.

Troubleshooting¶

Problem	Cause / Fix
"No growth areas yet"	Run the Growth Explorer for a state first (see Analyzing Growth Areas)
Source count is 0	No data for that state+city yet — run discovery anyway to trigger collection, or check your API keys
"VRM scrape started in background"	No VA listings existed for the state; refresh in ~1 minute and re-run discovery
Nothing happened on POST	At least one source must be checked; also confirm the growth area is resolved (check the URL has `growth_area_id`)
Results don't show rent	Expected for HUD/USDA/ATTOM/County — only VRM carries rents